Semi-structured Interviews

The survey questionnaire was followed by recorded semi-structured interviews. Semi- structured interviews were chosen because of the structure and flexibility involved in this form of data gathering. Although they are guided by a schedule containing interview questions, “the researcher is allowed to re-adjust the questions or ask other questions which are not in the schedule in order to clarify or elaborate on certain points to gain more information”, as suggested by Frankfort-Nachmias and Nachmias

73

(1996:234). The informants were given the freedom to express their views on a topic, despite the fact that it is a structured interview that is guided by prepared questions. In order for the interviewer to obtain detailed information to meet the purpose of the study, probing was used to encourage the respondents to give reasons for any views they hold.

The semi-structured interview process is used to explore specific questions as well as open ended questions. I had prepared interview questions before hand, but allowed the students to stray from the question at times to follow-up the question to allow the natural flow of the conversation between myself and student as long as it was relevant. I also probed when necessary to get more details or explanations. Through the individual interviews, I hoped to be able to probe individual attitudes and perceptions more deeply. I hoped that through the individual interviews, I would allow the participants to air their own personal opinions and attitudes without the obstruction from others. I chose not to conduct focus interviews in order to avoid students influencing each other.

Administration of interviews:

The procedure for interviews was the same as for questionnaires. I spoke to the head of the MBCHB isiXhosa programme for medical school who informed tutors teaching isiXhosa at medical school about my research. The head arranged for me to have a meeting with the students before requesting candidates for interviews so she could introduce me to their class and request other tutors to do the same and give me access to their students. Prior to the interviews, I met with the students and explained to them what their involvement in my study entailed and that the information they were to share with me would be kept confidential and would be used only for the study purposes. I also explained to them that they could withdraw at any time during data collection. All the participants agreed to participate in the study. The individual interviews involved 12 participants. An interview schedule containing nine open–ended questions was used (see Appendix B).

74

The participants were randomly selected from those that had shown their willingness to participate and were representative of all students in factors like age and gender. I asked them to choose the location where they would want interviews to be done in an “informal relaxed atmosphere” as Frankfort-Nachmias & Nachmias (1996:240) suggest. All of them preferred their rooms and small classrooms in medical school since most of the interviews were conducted on Friday afternoons. The interview took approximately 30-45 minutes and students were given incentives for their time. All the interviews were recorded with a small tape recorder placed in a relatively unobtrusive position. The students did not appear to be intimidated by the tape. The real names of the respondents were kept anonymous and codes were used to identify each interviewee. Each student was informed that s/he could withdraw at any point during the interview. I also explained to the students that participation was optional. 3.2.2. Classroom observation

While analysing the questionnaire data, which helped me highlight key issues, I realised I needed to explore further. In particular, I needed to shift away from a narrow focus on language attitudes to a broader focus on other issues such as second language acquisition and second language pedagogy. This can be examined through classroom observations.

For my observations, I chose to use the relatively less-structured observation. The reason for this attributed to Gall and Borg (1993) who said that the origins of the less- structured method lie in the anthropology tradition which aimed to explore the social meanings that underpin behaviour in natural social settings. Gall and Borg (1993) continue to say that there are numerous approaches to observational research. They explain the difference between more structured observation and less-structured. They say the two approaches originate in different academic traditions, and have different aims, purposes and procedures. I use a degree of structure through use of an observation checklist alongside my hand written notes. I designed my checklist based partly on previous research, my research questions and on understanding issues of second language pedagogy.

75

For the observation, I looked at the ways in which second language is taught, i.e. how and to what extent they use participation in their teaching, their pedagogical styles and to what extent the students seemed to be engaged in language learning. Overall, the observation assessed the effectiveness of each lesson to produce the required outcome, which I believe is the ability to teach language learning and acquisition. Good language learning and acquisition will be the predictor in how ready they are for communicating with patients. Therefore I used my checklist as a guide of noting the gaps and strengths of each style, and the notes I took allowed me to identify the teachers’ shortfall in the classroom. I also noted how teachers and students relate to each other before the actual teaching happened. I conducted four classroom observations.

The table below gives the observation checklist.

Table 2: Key areas of observation with pedagogy checklist Students’ attitudes

Class time

Lesson : clear learning and language objectives Materials

Collaborative activities planned for purposeful talk Learning scaffolded using L2 methods

Opportunities to use L2 Students’ behaviour

Appropriate grouping of learners Evidence of learner progress in lesson Data analysis:

Descriptive statistics served as the method of analysing the quantitative data. The qualitative data was analysed by using the questionnaire, interviews, observations and the manual as guides to identify themes that arose in the data.

76

This research used triangulation as its theoretical perspective, aiming at student’s attitudes, teaching and learning. It is a qualitative research study, dealing with individuals own accounts of their attitudes, motivations, and learning experiences. The selected number of participants inevitably cannot represent the whole medical students’ population learning isiXhosa. However, the findings of the study may legitimately be expected to contribute to knowledge about isiXhosa language learners in the UCT Medical Faculty.

77

Chapter Four: Data Analysis and Discussions

Data analysis and Discussions

In the section, I will present data and analyse it drawing on the data that was collected. My data is based on questionnaires that were administered to students, observations of classroom teaching and interviews with learners. In addition I have looked the manual that was used in the classes.

The following is an analysis and discussion of the survey questionnaire. The questionnaire can be found in Appendix A.

4.1. Background interpretation of the survey questionnaire

A total of 63 second year medical students completed the Communication Skills Attitude Scale (Rees, Sheard, & Davies, 2002). The questionnaire was used to assess individual student’s attitudes towards learning isiXhosa Communication Skills Course. The CSAS consisted of 27 questions using four point scale, from strongly agree, agree to strongly disagree and disagree. The Statistical Package for Social Sciences was used for data analysis. The percentages will be presented in the form of tables and charts. As shown in Table 1, the sample consisted of more females (34) than males (29). The average of age of the sample population is between 19-22 years.

Table 1: The distribution of students according to different variables

Age number Percentage

19-20 39 62% 21-22 22 35% 23 & unknown 2 3% Gender Male 29 46% Female 34 54% Languages Monolingual 9 13% Bilingual 29 44%

78 Multilingual 25 38% Schooling Type MOD C 32 51% DET 7 12% HOR 5 7% HOD 4 5% Non-South Africans 12 20% Unknown 3 5%

Figure 1: Age Category of respondents

As indicated in Figure 1 above, 29% of the respondents were 19 years old, 22% respondents are 21 yrs old. 33% respondents are 20yrs old. 10% respondents are 22 years old. 3% are 23 yrs old, while another 3% is unknown. The students are adults and they need special needs and requirements as learners. Also adult learners are goal driven in language learning. When looking back at theory about age, Brown, (2001:60) mentions that children and adults learning a second language tend to acquire that language differently. Children acquire new language subconsciously and naturally when it is delivered to them. Children often acquire the new language inductively and effortlessly, unlike adult second language learners. Adults tend to focus on parts of the language trying to comprehend how the new language works and how it is different from their mother tongue. They often tend to analyse language forms.

10% 22% 33% 29% 3% 3% 22 yrs 21yrs 20yrs 19yrs 23 Unknown

79

This, however, may impact negatively on their speed of language acquisition and, because of this, the adult learner tends to take more time to move away from focusing on forms of language to the purposes of the language. Hence it is important to nurture the learners’ because in early stages of learning a language, adult students are fragile and need support from their teachers. This is the time the teacher needs to exercise patience and empathy to maximise the development of their new language.

Figure 2: Gender

The female participants in the survey comprised 54% of the total research population, while male participants comprised 45%.

Figure 3: Language subjects speak 46% 54% Male Female 14% 46% 40% 0%

80

A total of 46% of the respondents considered themselves bilingual, which for the majority of them was English and their mother tongue. 40% of students considered themselves multilingual, which means they could speak more than two languages. Students who considered themselves to be monolingual totalled 14% of the research population, and because English is the university medium of instruction, it is safe to assume that it is English. When combining bilingual and multilingual percentages, one can see that majority of students are exposed to more than one language, which can be Afrikaans and English, English and isiXhosa and another language.

Schooling

The students came from different schooling backgrounds. Figure 4 shows the schooling background of the students. English is being taught as the medium of instruction in most of these schools, except the former Department of Education and Training (DET) schools. Former Model C Schools were previously under the control of the House of Assembly during the apartheid era and are usually ‘white schools’. Students from Model C schools constitute the largest group of students (51%). Model C schools were and still are the schools that have the best facilities, best teachers and best educational opportunities for children, followed by schools for non-South African students at (20%). Students from the former DET schools constituted 12%. DET schools are usually schools with little or no facilities which are mostly situated in townships, informal settlements and rural areas. The former House of Representatives (HOR) was the department that handled coloured children’s schooling and 7% of the survey group come from the HOR. Former HOR schools, although they were not quite as sidelined as DET schools, still have really poor infrastructure and facilities. House of Delegates (HOD) schools had been reserved for Indian pupils before 1994. Students from HOD constituted 5% of the survey population.

81

Language Statements

This section will look at two statements that were used in the survey to test students thinking in relation to language learning, and statements which are used to test language attitudes.

Statement 1: More should be done to promote indigenous languages in